A5005495160- Protein Structure and Dynamics
- Computational Drug Discovery Methods
- Polyomavirus and related diseases
- Machine Learning in Materials Science
- RNA Research and Splicing
- RNA regulation and disease
- Receptor Mechanisms and Signaling
- Advanced Proteomics Techniques and Applications
- Medical Imaging and Pathology Studies
- Mass Spectrometry Techniques and Applications
- Bacteriophages and microbial interactions
- Lipid Membrane Structure and Behavior
- Chemistry and Chemical Engineering
- Machine Learning in Bioinformatics
- Enzyme Structure and Function
- Neuroscience and Neuropharmacology Research
- Fuel Cells and Related Materials
- Bioinformatics and Genomic Networks
Acellera (Spain)
2018-2024
Helmholtz Zentrum München
2024
Universität Ulm
2024
Institute of Molecular Biology
2024
Center for Integrated Protein Science Munich
2024
Technical University of Munich
2024
Barcelona Biomedical Research Park
2018-2022
Universitat Pompeu Fabra
2017-2022
Hospital del Mar Research Institute
2017
Machine learning potentials have emerged as a means to enhance the accuracy of biomolecular simulations. However, their application is constrained by significant computational cost arising from vast number parameters compared with traditional molecular mechanics. To tackle this issue, we introduce an optimized implementation hybrid method (NNP/MM), which combines neural network potential (NNP) and mechanics (MM). This approach models portion system, such small molecule, using NNP while...
Structure-based drug discovery methods exploit protein structural information to design small molecules binding given pockets. This work proposes a purely data driven, structure-based approach for imaging ligands as spatial fields in target We use an end-to-end deep learning framework trained on experimental protein-ligand complexes with the intention of mimicking chemist's intuition at manually placing atoms when designing new compound. show that these models can generate images ligand...
Abstract For many decades virtual screening methods have provided a convenient and cost effective in silico solution the early stages of drug discovery. In particular, molecular docking uses structural information to approximate protein–ligand recognition, providing valuable for large chemical libraries at fast pace with multiple success stories validate approach. Nevertheless, turnaround results required assumptions approximations which compromise accuracy these algorithms. On other side...
Mutations in the human PURA gene cause neurodevelopmental syndrome. In contrast to several other monogenetic disorders, almost all reported mutations this nucleic acid-binding protein result full disease penetrance. study, we observed that patient across impair its previously co-localization with processing bodies. These either destroyed folding integrity, RNA binding, or dimerization of PURA. We also solved crystal structures N- and C-terminal PUR domains combined them molecular dynamics...
Mutations in the human PURA gene cause neurodevelopmental syndrome. In contrast to several other monogenetic disorders, almost all reported mutations this nucleic acid-binding protein result full disease penetrance. study, we observed that patient across impair its previously co-localization with processing bodies. These either destroyed folding integrity, RNA binding, or dimerization of PURA. We also solved crystal structures N- and C-terminal PUR domains combined them molecular dynamics...
Deep learning has been successfully applied to structure-based protein–ligand affinity prediction, yet the black box nature of these models raises some questions. In a previous study, we presented KDEEP, convolutional neural network that predicted binding given complex while reaching state-of-the-art performance. However, it was unclear what this model learning. work, present new application visualize contribution each input atom prediction made by network, aiding in interpretability such...
Machine learning (ML) is a promising approach for predicting small molecule properties in drug discovery. Here, we provide comprehensive overview of various ML methods introduced this purpose recent years. We review wide range properties, including binding affinities, solubility, and ADMET (Absorption, Distribution, Metabolism, Excretion, Toxicity). discuss existing popular datasets molecular descriptors embeddings, such as chemical fingerprints graph-based neural networks. highlight also...
Abstract The serotonin 5‐hydroxytryptamine 2A (5‐HT ) receptor is a G‐protein‐coupled (GPCR) relevant for the treatment of CNS disorders. In this regard, neuronal membrane composition in brain plays crucial role modulation functioning. Since cholesterol an essential component membranes, we have studied its effect on 5‐HT dynamics through all‐atom MD simulations. We find that presence increases conformational variability most segments. Importantly, detailed structural analysis indicates goes...
SkeleDock is a scaffold docking algorithm which uses the structure of protein–ligand complex as template to model binding mode chemically similar system. This was evaluated in D3R Grand Challenge 4 pose prediction challenge, where it achieved competitive performance. Furthermore, we show that if crystallized fragments target ligand are available then can outperform rDock software at predicting mode. Application Note also addresses capacity this macrocycles and deal with hopping. be accessed...
Abstract G protein-coupled receptors (GPCRs) are involved in numerous physiological processes and the most frequent targets of approved drugs. The explosion number new 3D molecular structures GPCRs (3D-GPCRome) during last decade has greatly advanced mechanistic understanding drug design opportunities for this protein family. While experimentally-resolved undoubtedly provide valuable snapshots specific GPCR conformational states, they give only limited information on their flexibility...
Mutations in the human PURA gene cause neuro-developmental syndrome. In contrast to several other mono-genetic disorders, almost all reported mutations this nucleic acid binding protein result full disease penetrance. study, we observed that patient across impair its previously co-localization with processing bodies. These either destroyed folding integrity, RNA or dimerization of PURA. We also solved crystal structures N- and C-terminal PUR domains combined them molecular dynamics...
Mutations in the human PURA gene cause neuro-developmental syndrome. In contrast to several other mono-genetic disorders, almost all reported mutations this nucleic acid binding protein result full disease penetrance. study, we observed that patient across impair its previously co-localization with processing bodies. These either destroyed folding integrity, RNA or dimerization of PURA. We also solved crystal structures N- and C-terminal PUR domains combined them molecular dynamics...
Abstract Mutations in the human PURA gene cause neuro-developmental syndrome. In contrast to several other mono-genetic disorders, almost all reported mutations this nucleic acid binding protein result full disease penetrance. study, we observed that patient across impair its previously co-localization with processing bodies. These either destroyed folding integrity, RNA or dimerization of PURA. We also solved crystal structures N- and C-terminal PUR domains combined them molecular dynamics...
Machine learning potentials have emerged as a means to enhance the accuracy of biomolecular simulations. However, their application is constrained by significant computational cost arising from vast number parameters compared traditional molecular mechanics. To tackle this issue, we introduce an optimized implementation hybrid method (NNP/MM), which combines neural network (NNP) and mechanics (MM). This approach models portion system, such small molecule, using NNP while employing MM for...
SkeleDock is a scaffold docking algorithm which uses the structure of protein-ligand complex as template to model binding mode chemically similar system. This was evaluated in D3R Grand Challenge 4 pose prediction challenge, where it achieved competitive performance. Furthermore, we show that, if crystallized fragments target ligand are available, can outperform rDock software at predicting mode. article also addresses capacity this macrocycles and deal with hopping. be accessed...